Exactly. When our car doesn't run well, a car with perhaps 20,000 parts, we take it to the mechanic who says "yup, issue was your spark plug, fixed". And even then, of course, the car issue may have been multifactorial. Maybe the spark plug failed because the fuel/air mix was off, for example, the spark plug was the symptom.
But then we leap from the car with 20,000 parts to a body with a trillion cells each cell with a trillion molecules. The calculus of the fuel/air mix and the spark plug has now been blown out of proportion, as challenging as conceiving a 20 dimensional manifold or the size of the universe. And so my reservation with a paper like this, and in general, is people say, "yup, sure was the spark plug" and then get accolades and a Nature paper, when the core issue was the air fuel mix, the air fuel mix that is, times a trillion cells, times a trillion molecules.
And I can't blame the authors. No shame in shooting for Nature and succeeding. No shame in these simplistic models, each one takes us a step further. But somewhere along the lines we're going to have think about the R&D of the big picture in non-hand-wavey ways.
The thing about a car is when the bumper gets detached you can't just strap it down to the front of the car and wait until it reattaches itself.
It may be (surely is) that something like lupus is multifactorial and impossibly complex, but knocking out the largest cause of something can be a reliably cure.
Look at antibitoics - a truly "replace the sparkplug" fix for let's say treating a MRSA infection. Doesn't treat cell damage, doesn't treat inflammation, doesn't treat pain, doesn't treat hormonal imbalance through the body. It doesn't even target the specific infection it kills indiscriminately. Yet "give them antibiotics until the MRSA goes away, or if it doesn't give them more antibiotics until it does" it super hand-wavy.
Sure, a car has a lot of parts, but when one of them is leaking oil where it's not supposed to, you don't have to check every part to see where the problem is.
Likely nanotechnology. Sensors that can reside in your body and give up to the moment details, and chemical factories that we can programmatically make adjustments to the body's existing pathways (reprogramming the immune system, etc.).
In other words, better integration and faster feedback loops.
When dealing with code we say refactor and simplify. Superintelligence my (rightfully) consider us too complicated and a 'big ball of mud' and replace us with version 2.0.
But then we leap from the car with 20,000 parts to a body with a trillion cells each cell with a trillion molecules. The calculus of the fuel/air mix and the spark plug has now been blown out of proportion, as challenging as conceiving a 20 dimensional manifold or the size of the universe. And so my reservation with a paper like this, and in general, is people say, "yup, sure was the spark plug" and then get accolades and a Nature paper, when the core issue was the air fuel mix, the air fuel mix that is, times a trillion cells, times a trillion molecules.
And I can't blame the authors. No shame in shooting for Nature and succeeding. No shame in these simplistic models, each one takes us a step further. But somewhere along the lines we're going to have think about the R&D of the big picture in non-hand-wavey ways.